add GET, DELETE responses API endpoints

litellm/proxy/common_request_processing.py

@@ -108,7 +108,13 @@ class ProxyBaseLLMRequestProcessing:
         user_api_key_dict: UserAPIKeyAuth,
         proxy_logging_obj: ProxyLogging,
         proxy_config: ProxyConfig,
-        route_type: Literal["acompletion", "aresponses", "_arealtime"],
+        route_type: Literal[
+            "acompletion",
+            "aresponses",
+            "_arealtime",
+            "aget_responses",
+            "adelete_responses",
+        ],
         version: Optional[str] = None,
         user_model: Optional[str] = None,
         user_temperature: Optional[float] = None,
@@ -178,7 +184,13 @@ class ProxyBaseLLMRequestProcessing:
         request: Request,
         fastapi_response: Response,
         user_api_key_dict: UserAPIKeyAuth,
-        route_type: Literal["acompletion", "aresponses", "_arealtime"],
+        route_type: Literal[
+            "acompletion",
+            "aresponses",
+            "_arealtime",
+            "aget_responses",
+            "adelete_responses",
+        ],
         proxy_logging_obj: ProxyLogging,
         general_settings: dict,
         proxy_config: ProxyConfig,

litellm/proxy/response_api_endpoints/endpoints.py

@@ -106,8 +106,49 @@ async def get_response(
     -H "Authorization: Bearer sk-1234"
     ```
     """
-    # TODO: Implement response retrieval logic
+    from litellm.proxy.proxy_server import (
-    pass
+        _read_request_body,
+        general_settings,
+        llm_router,
+        proxy_config,
+        proxy_logging_obj,
+        select_data_generator,
+        user_api_base,
+        user_max_tokens,
+        user_model,
+        user_request_timeout,
+        user_temperature,
+        version,
+    )
+
+    data = await _read_request_body(request=request)
+    processor = ProxyBaseLLMRequestProcessing(data=data)
+    try:
+        return await processor.base_process_llm_request(
+            request=request,
+            fastapi_response=fastapi_response,
+            user_api_key_dict=user_api_key_dict,
+            route_type="aget_responses",
+            proxy_logging_obj=proxy_logging_obj,
+            llm_router=llm_router,
+            general_settings=general_settings,
+            proxy_config=proxy_config,
+            select_data_generator=select_data_generator,
+            model=None,
+            user_model=user_model,
+            user_temperature=user_temperature,
+            user_request_timeout=user_request_timeout,
+            user_max_tokens=user_max_tokens,
+            user_api_base=user_api_base,
+            version=version,
+        )
+    except Exception as e:
+        raise await processor._handle_llm_api_exception(
+            e=e,
+            user_api_key_dict=user_api_key_dict,
+            proxy_logging_obj=proxy_logging_obj,
+            version=version,
+        )
 
 
 @router.delete(
@@ -136,8 +177,49 @@ async def delete_response(
     -H "Authorization: Bearer sk-1234"
     ```
     """
-    # TODO: Implement response deletion logic
+    from litellm.proxy.proxy_server import (
-    pass
+        _read_request_body,
+        general_settings,
+        llm_router,
+        proxy_config,
+        proxy_logging_obj,
+        select_data_generator,
+        user_api_base,
+        user_max_tokens,
+        user_model,
+        user_request_timeout,
+        user_temperature,
+        version,
+    )
+
+    data = await _read_request_body(request=request)
+    processor = ProxyBaseLLMRequestProcessing(data=data)
+    try:
+        return await processor.base_process_llm_request(
+            request=request,
+            fastapi_response=fastapi_response,
+            user_api_key_dict=user_api_key_dict,
+            route_type="adelete_responses",
+            proxy_logging_obj=proxy_logging_obj,
+            llm_router=llm_router,
+            general_settings=general_settings,
+            proxy_config=proxy_config,
+            select_data_generator=select_data_generator,
+            model=None,
+            user_model=user_model,
+            user_temperature=user_temperature,
+            user_request_timeout=user_request_timeout,
+            user_max_tokens=user_max_tokens,
+            user_api_base=user_api_base,
+            version=version,
+        )
+    except Exception as e:
+        raise await processor._handle_llm_api_exception(
+            e=e,
+            user_api_key_dict=user_api_key_dict,
+            proxy_logging_obj=proxy_logging_obj,
+            version=version,
+        )
 
 
 @router.get(

litellm/proxy/route_llm_request.py

@@ -22,6 +22,8 @@ ROUTE_ENDPOINT_MAPPING = {
     "amoderation": "/moderations",
     "arerank": "/rerank",
     "aresponses": "/responses",
+    "aget_responses": "/responses",
+    "adelete_responses": "/responses",
 }
 
 
@@ -48,6 +50,8 @@ async def route_request(
         "arerank",
         "aresponses",
         "_arealtime",  # private function for realtime API
+        "aget_responses",
+        "adelete_responses",
     ],
 ):
     """

litellm/router.py

@@ -3083,6 +3083,12 @@ class Router:
             "responses",
             "afile_delete",
             "afile_content",
+            # responses api GET
+            "get_responses",
+            "aget_responses",
+            # responses api DELETE
+            "delete_responses",
+            "adelete_responses",
         ] = "assistants",
     ):
         """
@@ -3130,6 +3136,10 @@ class Router:
             elif call_type in (
                 "anthropic_messages",
                 "aresponses",
+                "get_responses",
+                "aget_responses",
+                "delete_responses",
+                "adelete_responses",
             ):
                 return await self._ageneric_api_call_with_fallbacks(
                     original_function=original_function,